READMEReplication Files and Data for:Julie Anne Weaver�Electoral Dis-connection: The Limits of Re-Election in Contexts of Weak Accountability� Journal of PoliticsThere are two separate files with code to reproduce all of the analysis, tables and plots, one for the main text and one for the online appendix:1. ReplicationScript_JulieAnneWeaver_ElectoralDisconnection_MainText.R2. ReplicationScript_JulieAnneWeaver_ElectoralDisconnection_Appendix.RThere are five datasets:1. conjoint.data: Dataset to reproduce the results of the conjoint experiment (main text and appendix).  Unit of observation is the candidate profile seen by a given respondent.  7576 observations.2. RDD.data: Dataset for the RDD results (main text and appendix).  Unit of observation is the candidate.  Includes all observations (i.e. not just those within the bandwidth used in the paper) across 3 election years (2006, 2010 and 2014).  34,737 observations.3. RDD.data.heterogtxt: Dataset for the heterogonous treatment effects design (main text and appendix).  Unit of observation is the candidate.  Includes observations across 3 election years (2006, 2010 and 2014), but only within the bandwidth used in the paper.  8,981 observations.4. RDD.data.heterogtxt.district: Same as RDD.data.heterogtxt except only includes municipal-level data per year (rather than candidate-level data).  Is only used to test the correlation between trust in horizontal oversight institutions and mayoral performance (main text). Unit of observation is the municipality.  4,912 observations.5. survey.data: Dataset from the original survey, excluding the conjoint experiment (main text and appendix).  Unit of observation is the respondent.  1,061 observations.